Sixty percent of all cancers occur in men and women aged 65 or older at diagnosis. A number of investigators have shown that older individuals with cancer are at increased risk for receiving less than definitive treatment for cancer, and that such non-definitive treatment contributes to poor survival. This proposal will examine the use of chemotherapy in older men and women with colon cancer and older women with breast cancer. Based on the results of numerous clinical trials, the NCI consensus conferences and other authorities recommend adjuvant chemotherapy for all men and women with stage III colon cancer, and for women up to age 70 with breast cancer > 1.0 cm in size. Nevertheless, others and we have found that fewer than half such patients in the community actually receive chemotherapy. In addition, there are substantial variations in use of chemotherapy by geographic region and also by patient characteristics, after controlling for tumor characteristics such as size, stage, hormone receptor status and histologic grade. Finally, others and we have found that only about half of older cancer patients see a medical oncologist. Thus we propose a two steps model governing receipt of chemotherapy: one set of factors influences whether or not the patient sees a medical oncologist and another set of factors influences whether the patient, once she or he sees an oncologist, actually receives chemotherapy. We now propose to use the SEER/Medicare linked data to address the following aims: determine the factors influencing whether older colon and breast cancer patients with tumor characteristics for which chemotherapy is generally recommended actually see a medical oncologist; and determine the factors influencing receipt of chemotherapy by the patients who see a medical oncologist. These analyses will be performed at the level of the patient and at the level of the provider. This will allow for the generation of provider-level information on an important process of quality cancer care in the elderly. We will then explore the reliability of the provider-level data in order to assess its potential utility in assessing quality of care for individual providers.